Over the past several days a number of thoughtful questions arrived on my page, and I answered each one as it came. Several of them, though, belong together, because they are really one question wearing different clothes: can we trust the text of the New Testament at all? I want to set out the case in order, with the data behind it, beginning with the objection I hear most often. “The originals were never found.”
Every Ancient Text Starts in the Same Place
The objection is true, and it proves far less than people think. No autograph survives for any work of classical antiquity. Not Homer, not Caesar, not Tacitus. If the absence of an original disqualified a text, we would have to discard the entire library of the ancient world and most of its history with it.
Scholars do not ask whether the original exists. They ask how well the surviving copies allow us to recover it. The discipline that does this work is called textual criticism, and it evaluates any tradition by three measurable quantities: how many manuscripts survive, how much time passed between composition and the earliest surviving witness, and how closely independent lines of transmission agree with one another. Those are the tests. They are the same tests for the Gospels as for the Gallic Wars.
The New Testament Against Its Rivals
Begin with volume. The New Testament survives in approximately 5,800 catalogued Greek manuscripts. Using Aland’s classification, that includes 128 papyri, roughly 320 majuscules, about 2,900 minuscules, and some 2,400 lectionaries. Add the early translations and the number grows again: an estimated 10,000 Latin manuscripts and another 5,000 to 10,000 witnesses in Syriac, Coptic, Armenian, Georgian, and Ethiopic.
Now the competition.
- Homer’s Iliad, the closest rival in antiquity, survives in fewer than 2,000 manuscripts.
- Caesar’s Gallic Wars survives in roughly 250, and the earliest copy we possess was made about 900 years after Caesar wrote.
- Tacitus wrote the Annals in the second century. Two principal manuscripts carry the text, one from the ninth century and one from the eleventh.
- Herodotus and Thucydides each come to us in roughly eight manuscripts, with gaps between composition and copy exceeding 1,300 years.
Nobody doubts that we possess what Caesar wrote. Yet on volume alone the New Testament outstrips him by an order of magnitude, and the gap only widens when we turn to the question of time.
How Close to the Source
Volume means little if every copy is late. So consider the interval. The earliest New Testament witness is P52, the Rylands Greek Papyrus 457, a fragment containing John 18:31-33 and 37-38 and dated on paleographic grounds to the second century. Behind it stand P46, which carries the Pauline corpus, and P66 and P75, which carry John, all falling roughly between AD 175 and 225.
P75 deserves a special word. When it is compared with Codex Vaticanus, a fourth-century manuscript, the two agree to a degree that demonstrates careful copying across an interval of about 150 years. That is not what a careless or inventive tradition looks like.
Why More Copies Means More Accuracy
Here my training in cryptography and information theory becomes useful, because the skeptic’s picture of transmission is exactly backwards. He imagines a single chain of copies, each adding its errors to the last, until the message is hopelessly degraded. That is not how the New Testament was transmitted, and it is not how information behaves.
When a message is copied into many channels that are separated geographically and linguistically, a corruption that enters one channel does not replicate in the others. Compare the channels and the error stands out: you can see where it occurred and what the original reading was. This is the operating principle behind every error-correcting code in modern communications. Redundancy does not degrade a signal. It is precisely what makes a signal recoverable. And the New Testament tradition is the most redundant text in the ancient world by an order of magnitude.
What About the Variants?
You will hear that there are some 400,000 variants among the manuscripts, and the number is roughly right. What is left unsaid is what those variants are. The overwhelming majority are matters of spelling, word order, or one synonym in place of another, differences that vanish entirely in translation. They are the fingerprints of thousands of human copyists, and they are exactly what the comparison method is designed to resolve.
Two passages of any length are genuinely disputed on manuscript grounds: Mark 16:9-20 and John 7:53 to 8:11. Both are footnoted in every serious edition. Beyond those, the residue is small, and no doctrine of the faith depends on a contested reading. Bart Ehrman, among the most cited critics of Christian origins, concedes this very point in the appendix to his own popular book on textual variation.
Is This Circular Reasoning?
A second objection often follows the first: “You are using the Bible to prove the Bible.” Circular reasoning would be an appeal to Scripture’s authority in order to establish Scripture’s authority. Nothing above does that.
A manuscript count is an inventory, nothing more. Paleographic dating is a physical method, applied by specialists who have no confessional stake in the outcome. The witnesses themselves are objects you can go and see. Codex Sinaiticus sits in the British Library. Codex Vaticanus is in the Vatican Library. The Bodmer papyri are in Cologny and the Vatican, and the Rylands fragment is in Manchester. The argument built on them is an argument from external evidence, measured by criteria applied uniformly to every ancient text. If those criteria are good enough for Caesar, they are good enough for John.
I do not expect data alone to make anyone a believer. But I do expect it to clear away a bad reason for unbelief, and “the originals were never found” is a bad reason. The originals of everything are gone. What survived, in the case of the New Testament, is a tradition of astonishing depth that lets us read what the apostles wrote. The prophet said it long before any of this could be measured:
The grass withereth, the flower fadeth: but the word of our God shall stand for ever. (Isaiah 40:8)
In the next article I turn to a different objection: that the Gospels were not written by eyewitnesses, and that Galilean fishermen could not have written Greek in any case.
Get new articles by email
Subscribe and I will send you a short note whenever a new article is published on this site. No spam, and you can unsubscribe at any time.